Development and Country Applications of a Program Planning and Monitoring and Evaluation Tool for Oral Pre-exposure Prophylaxis
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
WHITE PAPER
Development and Country
Applications of a Program
Planning and Monitoring and
Evaluation Tool for Oral
Pre-exposure ProphylaxisAUTHORS
Avenir Health
Nicole M. Bellows
Lori Bollinger
Matthew Hamilton
Katharine Kripke
FHI 360
Tiffany Lillie
Jill Peterson
Kristine Torjesen
USAID
Lopa Basu (USAID Vietnam)
Delivette Castor*
Robyn Eakle
Chris Obermeyer**
Trang Ngo (USAID Vietnam)
Sangeeta Rana
Country partners
Micah Anonya, JHPIEGO, Kenya
Noela Chicuecue, National STI-HIV/AIDS Control Program, Ministry of Health, Mozambique
Sibongile Maseko, Eswatini
Sindy Matse, Ministry of Health, Eswatini
Getrude Ncube, Ministry of Health and Child Care, Zimbabwe
Definate Nhamo, Pangaea Zimbabwe AIDS Trust, Zimbabwe
Kwadwo Koduah Owusu, National AIDS/STI Control Programme, Ghana
Regeru Njoroge Regeru, LVCT Health, Kenya***
*Currently at Columbia University
**Currently at FHI 360
***Currently at Johnson & Johnson Global Public Health
ACKNOWLEDGMENTS
The authors would like to acknowledge Anaise Kanimba, Joseph Murungu, Nuno Gaspar, Massai
Belayneh, and Meghan O’Keefe Douglas for their support in the development of PrEP-it.
This work was made possible by the generous support of the American people through the U.S.
Agency for International Development (USAID) and the U.S. President’s Emergency Plan for AIDS Relief
(PEPFAR). It is the result of a collaboration between the Optimizing Prevention Technology Introduction
On Schedule (OPTIONS) Consortium and the Collaboration for HIV Prevention Options to Control the
Epidemic (CHOICE). Financial assistance was provided by USAID (https://www.usaid.gov/) to FHI 360
under the terms of Cooperative Agreement No. AID-OAA-A-15-00035 and Cooperative Agreement No.
7200AA19CA00002. The contents of this document are those of the authors and do not necessarily
reflect the views of USAID or the U.S. Government.
1TABLE OF CONTENTS
ACRONYMS ........................................................................................................................... 3
EXECUTIVE SUMMARY ........................................................................................................... 4
INTRODUCTION ..................................................................................................................... 4
PrEP-IT DEVELOPMENT .......................................................................................................... 5
Development Process ........................................................................................................ 5
Findings ............................................................................................................................. 6
PrEP-IT STRUCTURE AND FEATURES ....................................................................................... 8
COUNTRY EXPERIENCES USING PrEP-IT ................................................................................ 12
Eswatini........................................................................................................................... 12
Zimbabwe ....................................................................................................................... 12
Ghana.............................................................................................................................. 13
Vietnam........................................................................................................................... 13
LESSONS LEARNED............................................................................................................... 13
CONCLUSION ............................................................................. 1Error! Bookmark not defined.
REFERENCES ........................................................................................................................ 16
2ACRONYMS
3TC lamivudine
AGYW adolescent girls and young women
AIM AIDS Impact Model
ARV antiretroviral
ART antiretroviral therapy
FSW female sex worker
FTC emtricitabine
M&E monitoring and evaluation
MSM men who have sex with men
PEPFAR U.S. President’s Emergency Plan for AIDS Relief
PrEP pre-exposure prophylaxis
PrEP-it PrEP Implementation, Planning, Monitoring, and Evaluation Tool
PWID people who inject drugs
SDC serodifferent couple
TDF tenofovir disoproxil fumarate
UNAIDS Joint United Nations Programme on HIV/AIDS
USAID United States Agency for International Development
WHO World Health Organization
3TOC
EXECUTIVE SUMMARY
Governments, donors, and other stakeholders in countries at the forefront of delivery for oral pre-
exposure prophylaxis (PrEP) have encountered challenges in setting targets, estimating costs, and
tracking PrEP delivery. In response, the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) funded
the development of a Microsoft Excel-based tool called the PrEP Implementation, Planning, Monitoring,
and Evaluation Tool, or “PrEP-it.” This paper describes the development of PrEP-it, lessons learned
during development, and the application of PrEP-it for target-setting in Eswatini and Zimbabwe and drug
supply management in Ghana and Vietnam.
PrEP-it development occurred from October 2018 through September 2019. The development stages
included: (1) a literature review to serve as a knowledge foundation and identify research gaps; (2) the
assembly of national data sources to assist with tool inputs; (3) information-gathering visits to
Mozambique and Zimbabwe; (4) consultation calls with PrEP experts; (5) tool development; (6) pilot
testing in Eswatini, Kenya, Mozambique, and Zimbabwe; and (7) tool revision and release in September
2019. Key data sources used in the tool development include HIV prevalence estimates from Spectrum’s
AIDS Impact Model (AIM), PrEP costing studies from six countries, and Spectrum’s Goals Model outputs
on PrEP impact. In country applications, PrEP-it was used as an organizing framework to help
stakeholders assemble data and make key decisions required for target-setting and commodity
forecasting.
The primary need was for a tool to assist with target-setting, estimating capacity, forecasting costs and
epidemiological impact, tracking program delivery, and projecting future client flow and commodity
needs. Testing the tool with country-level stakeholders was critical in finding the right balance between
flexibility, specificity, and user-friendliness in PrEP-it development. The calculations for targets, costs,
and impact are all dependent on the continuation rates of populations on PrEP at months 1, 3, 6, and 12
after initiation. Country applications of PrEP-it demonstrated the importance of incorporating
continuation rates in setting targets and forecasting drug supply needs, resulting in lower projected
costs and drug needs than originally expected.
PrEP-it advances prior approaches to program planning, monitoring, and evaluation for PrEP by
performing multiple complementary functions and incorporating PrEP continuation and implementation
scale-up patterns, which had not previously been considered in target-setting and costing. A
forthcoming software version of PrEP-it will enhance successful scale-up of PrEP in low- and middle-
income countries.
INTRODUCTION
In 2015, the World Health Organization recommended oral PrEP for individuals at substantial risk of HIV
infection who may not consistently use condoms.1 Since then, more than 80 countries have introduced
ongoing PrEP delivery for those at substantial risk of HIV infection.2 In sub-Saharan Africa, where 70% of
the people accessing antiretroviral (ARV) drugs for HIV treatment reside, PrEP programs have focused
primarily on serodifferent couples (SDCs), female sex workers (FSWs), men who have sex with men
(MSM), and adolescent girls and young women (AGYW).3 Some countries have identified additional
priority populations for PrEP provision, such as truck drivers and incarcerated persons.4, 5
4PrEP implementation has varied both within and among countries in terms of service delivery models
and populations of focus. While these variations allow adaptations to different contexts, they also pose
challenges in planning and monitoring and evaluation (M&E) of programs supported by large donors
such as PEPFAR, the Global Fund to Fight AIDS, Tuberculosis and Malaria, and host national
governments. These stakeholders are often tasked with establishing targets for how many people
should be reached with PrEP in a given time period.
Rapid scale-up of PrEP is one of the five prevention pillars for achieving the Joint United Nations
Programme on HIV/AIDS (UNAIDS) goals of ending the HIV/AIDS epidemic; however, gaps in
demographic, epidemiological, and behavioral data have impeded efforts to generate evidence-based
targets.6, 7 The uncertainties around the continuation of PrEP after initiation and the expectations for
scale-up make annual target-setting and forecasting particularly challenging, with the risk that
programs’ assumptions about high continuation rates could result in overestimates of drug supply
needs.8, 9
During a PrEP mathematical modeling exercise conducted in 2017 in collaboration with stakeholders in
Eswatini, Lesotho, Mozambique, and Uganda,10 country stakeholders expressed a need for assistance
with target-setting, costing, and monitoring of PrEP programs. In response, PEPFAR and the U.S. Agency
for International Development (USAID) funded the development of an innovative multi-module PrEP
Implementation Planning, Monitoring, and Evaluation Tool (PrEP-it) to support setting targets,
estimating capacity, forecasting costs and impact, tracking program delivery along the PrEP initiation
cascade, and projecting future client flow and commodity needs.
This paper describes the development, structure, and features of PrEP-it and the lessons learned during
the development and initial applications of the tool.
PrEP-IT DEVELOPMENT
Development Process
Most of the PrEP-it development occurred from October 2018 through September 2019. The
development process included a literature review to provide a knowledge foundation; collation of
national data sources for 35 countries and regional programs to assist with data inputs; information-
gathering visits to Mozambique and Zimbabwe; global consultations with PrEP experts; tool
development; pilot visits to Eswatini, Kenya, Mozambique, and Zimbabwe; and tool revision and release.
From October through December 2018, members of the PrEP-it team conducted a literature review to
assess what was known about PrEP implementation in scaled-up settings, including target-setting,
monitoring, and costing. We searched the peer-reviewed literature in the PubMed, Popline, and Global
Health databases using keywords aimed at PrEP implementation, such as targets, costs, and scale-up. In
addition, we reviewed abstracts from recent (2016–2018) International AIDS Society and HIV Research
for Prevention conferences, as well as the prepwatch.org website and the AVAC database of current,
planned, and completed PrEP demonstration projects, national guidelines, and implementation
initiatives.2, 11
Because PEPFAR/USAID emphasized a need for default values, the team assembled several data sources
(Table 1) to inform estimates of the number of potential PrEP users for 35 countries/regions. In addition,
we analyzed cost data from PrEP programs in six sub-Saharan Africa countries12–18 and combined them
5with country-specific salary data19, 20 to generate default values for visit costs. Finally, we used outputs
from the Goals Model21 for impact factors, expressed as HIV infections averted per person-year of PrEP.
Table 1. Data Sources
Module Data Sources
Target-setting United Nations population estimates (2018)
UNAIDS report 2018
Key Populations Atlas
HIV prevalence estimates (AIM 2018)
Various peer-reviewed literature sources
Most recently available Demographic and Health Survey data
Costs Country-specific salary data
PrEP reference cost studies from six countries (Eswatini, Kenya,
Lesotho, South Africa, Zambia, Zimbabwe)
Impact Goals Model outputs
The PrEP-it team also held consultations with country stakeholders in Mozambique and Zimbabwe,
including Ministry of Health officials, PEPFAR country teams, and PEPFAR partners implementing PrEP
programming. Information was collected on the priority populations being served, implementation
approaches, health information systems, and support needs for program planning and M&E. In
Mozambique, team members visited two facilities implementing PrEP services to review data collection
tools and gain further insight into different implementation models.
To obtain global input on the tool, the PrEP-it team held a series of nine remote consultations with 32
experts from organizations involved in PrEP policy, research, or delivery. The stakeholders discussed
important assumptions to incorporate in PrEP-it, such as the most common M&E metrics, scale-up
patterns, and measures for estimating elevated likelihood for HIV acquisition.
The tool was field-tested in Eswatini, Kenya, Mozambique, and Zimbabwe. These field tests included
holding meetings and workshops with Ministry of Health officials, PEPFAR teams, and PEPFAR
implementing partners delivering PrEP, reviewing how the tool worked, and soliciting feedback on how
it could be improved.
After revision, PrEP-it was released in September 2019, along with a user guide, introductory webinar,
and instructional videos. In March 2020, we added a module to PrEP-it that allows users to identify
localities that should be prioritized for PrEP provision to AGYW based on cost-effectiveness. Since the
initial tool release, further updates to the modules have been made based on user feedback.
Findings
The literature review yielded a repository of 311 references, including 10 from peer-reviewed sources
and eight from conference abstracts. The remaining references consisted of global and national
guidelines, project descriptions from the AVAC database, and reports from prepwatch.org. Studies
6suggested relatively high uptake of PrEP and a high discontinuation rate, with up to half of those
initiating not returning for a one-month visit due to side effects, stigma, negative attitudes among
health care workers, limited access to clinics, and tedious refill processes.22–24 Reinitiation of PrEP was
also observed, with one study finding 17% of
AGYW restarting PrEP within three months of The review confirmed the need for
discontinuation.25 While useful, the review interaction with country partners and
identified many gaps in the literature, confirming global experts to ensure the tool would
the need for interactions with country partners
help the intended users fill critical gaps in
and global experts to ensure the tool would help
knowledge and planning practice.
the intended users fill critical gaps in knowledge
and planning practice.
The initial consultations with country policy makers and program implementers highlighted specific
challenges that PrEP-it could address for two primary audiences. First, it could serve a program planning
function for governments and donors, to help them set targets and consider the corresponding coverage
of potential users, estimated costs, and expected impact in terms of the number of HIV infections
averted. Second, it could help implementers assess capacity for the delivery of PrEP, track the efficiency
of the PrEP initiation cascade,26 and project future client flow based on initiation and continuation rates.
In Mozambique, participants were interested in the use of PrEP-it for target-setting, particularly for
annual PEPFAR planning. Participants in Zimbabwe expressed a need for assistance in estimating
patterns of PrEP use, such as continuation rates, re-initiation, and episodic use. In response, a
Continuation Calculator was included, allowing users to input client-level data to estimate continuation
rates and the percentage of clients reinitiating during the data collection period.
The global consultations yielded several contributions that guided tool development. To maximize
potential utility, experts recommended that the tool be designed to work at any level, including at the
national, regional, implementing partner, and site levels. This flexibility means that in Microsoft Excel,
the tool cannot be set up to allow for geographic breakdowns for most of the inputs and outputs, such
as tracking the PrEP initiation cascade and drug forecasting. However, some components of the tool do
allow users to conduct sublocation analyses, including the Capacity module, the AGYW Geographic
Prioritization module, and the function enabling them to disaggregate targets by district.
Experts also emphasized the diversity of approaches that programs use to implement and manage PrEP
and the differing access to data for data inputs, noting that some users would need suggested default
values for population size estimates, continuation rates, and costs. The tool accommodates those
differences by allowing a smaller set of required data inputs for less complex analyses. Furthermore,
while a Detailed Cost module allows for a rigorous costing approach, a “Costs Lite” module with less
detail was also included to allow for simpler cost analyses.
Feedback from the field tests provided several suggestions for clarifications, adjustments, and
improvements. Throughout the consultation and tool development process, ease of use was
underscored as a priority. The team included several features to improve the user interface, including
color-coding and a navigational panel to direct users based on their specific objectives. In the Capacity
module, the way drug capacity was entered for each site was altered to provide an algorithm for
allocating national supply to sites.
7PrEP-IT STRUCTURE AND FEATURES
PrEP-it can be used at any level (e.g., national, provincial, district, country, facility, or implementing
partner) and users can select which functions of the tool they want to use. Required inputs and available
outputs for each of the seven functions are summarized in Table 2. In addition, the tool identifies the
minimum inputs needed to interact with the rest of the modules and to access additional functionality.
The most prominent feature of PrEP-it is the breadth of functions available, addressing many aspects of
PrEP planning and service delivery.
By addressing and linking target-setting, capacity, the initiation cascade, costing, drug supply, and
impact in one tool, PrEP-it greatly expands the number and type of available outputs in comparison with
tools that examine only one or two of these dimensions. A critical feature of the tool is that all
calculations within the tool are based on continuation rates, i.e., the percentages of initiating clients
who continue to take PrEP at months 1, 3, 6, and 12 after initiation. Continuation rates are used in
target-setting to calculate the number of persons on PrEP each month, which then feeds into estimates
of PrEP coverage, costs, impact, and drug supply needs.
Table 2. PrEP-it Functions
Minimum Additional Inputs Within-Module
Inputs Outputs
Monthly initiation Monthly tracking of additional data related to Summary of PrEP cascade
visits uptake and other recommended PrEP indicators from HIV testing to
initiation
At least 1 Allows up to 6 different continuation curves
CONTINUUM
continuation Forecasts for the coming
curve (% clients Client-level data to calculate continuation curves months based on recent
continuing PrEP at past initiations and
months 1, 3, 6, continuation rates
and 12 after
initiation without
a gap in service)
Overall monthly If overall capacity is not known, capacity is Calculations of overall
capacity for calculated based on sub-unit (e.g., site-level) data capacity
initiation and on:
continuation visits Bottlenecks identified for
• Staffing capacity (up to 2 staff types) each sub-unit
CAPACITY
• Laboratory capacity Efficiency analyses to
maximize resource
• Aggregated antiretroviral (ARV) supply allocation
• Potential PrEP clients (e.g., monthly
demand for PrEP)
8Minimum Additional Inputs Within-Module
Inputs Outputs
Priority For targets based on coverage: Annual and monthly
populations • Priority population size targets per priority
population
Target time • HIV prevalence
period Targets can be
TARGET-SETTING
• Percentage of each priority population at disaggregated by
Continuation high risk sublocation, age, and sex
curve(s) and
scale-up • Percentage desired to be taking PrEP at the
pattern(s) end of the target period
Previously For targets based on capacity:
established • Complete Capacity module
targets
Annual visit Detailed Cost module allows for further specification Cost per person initiated
schedule(s) of costs: on PrEP, accounting for
discontinuation
Monthly ARV and • More detailed staffing data, including task-
adherence shifting analyses Cost per person on PrEP
support costs for a full year
COSTS
• Service delivery models (e.g., fixed vs.
Initiation and mobile, urban vs. rural, different
continuation visit populations within a site, different
costs using Costs geographic areas)
Lite module
• Greater specification of lab schedules
Above-site costs
• Pre-initiation visit costs
Selected Adjust defaults for PrEP efficacy, effective use, and Identified areas for
threshold annual discount rate geographic prioritization
criterion
Cost per infection averted threshold Total number of AGYW
HIV prevalence and HIV prevalence in
for each HIV prevalence among adolescent girls and young selected areas
AGYW GEOGRAPHIC
geographic area women (AGYW) ages 10–14 by geographic area
PRIORITIZATION
Total PrEP costs and total
Population AGYW and number initiating PrEP by cost savings for AGYW in
geographic area selected areas
Continuation rates by geographic area Total cost savings across
all areas
Delivery strategy by geographic area (based on
Detailed Cost module) Additional cost savings
achieved by targeting
Costs per person-year on antiretroviral therapy
(ART)
9Minimum Additional Inputs Within-Module
Inputs Outputs
No required Specify which ART/voluntary medical male No impact-only outputs
inputs circumcision scale-up scenario to use for impacts
IMPACT
Assign impact factors for custom populations
Adjust impact factors to account for selection of
higher-risk individuals within one of the listed
populations
Number of pills No additional inputs Monthly pills needed and
DRUG SUPPLY
distributed at associated costs for ARVs
initiation and associated with
continuation visits projections based on
program implementation
Cost per pill data or targets
CROSS-MODULE OUTPUTS
Continuum + Capacity = View potential capacity gaps based on service delivery forecasts
Continuum + Targets = Compare service delivery with targets
Continuum + Costs = Estimate costs associated with service delivery to see if costs are in line with expectations
Continuum + Impact = Estimate the number of HIV infections averted based on service delivery
Continuum + Targets = Set targets based on capacity for service delivery
Continuum + Targets = View potential capacity gaps with established targets
AGYW + Targets = Generate targets based on selection of areas with potential for most impact
AGYW + Impact = Identify more accurate impact factor for AGYW in selected geographic areas
Costs + Targets = Estimate costs based on targets
Impact + Targets = Estimate number of HIV infections averted associated with targets
Costs + Impact + Targets/Cascade = Estimate costs per HIV infection averted
The target-setting functions in PrEP-it allow for three different approaches to calculating or examining
targets for initiations: (1) use the Microsoft Excel Solver add-in to set targets by the desired proportion
of the eligible population taking PrEP at the end of the target-setting period based on the continuation
rates and scale-up patterns; (2) set targets by allocating health system capacity among different priority
populations; or (3) enter previously established targets to estimate population-level coverage and
produce outputs in conjunction with other modules related to capacity, costs, drug supply, and impact.
For all three approaches, users must define the priority populations (for example, SDCs, MSM), enter the
continuation rates for each priority population, and specify the scale-up pattern anticipated for rolling
out PrEP in each priority population.
PrEP impact is calculated via impact factors, which are the numbers of HIV infections averted over five
years per person-year of PrEP, for each indicated population. Default impact factor values were derived
for each of the 35 countries and for each population using the Spectrum Goals Model, a dynamic
compartmental model of HIV transmission by risk group.27 PrEP efficacy was assumed to be 90%, and
adherence was assumed to be 100%. We assumed 50% coverage in 2020 for each population and output
HIV infections averted over the five-year period from 2020 to 2024. We then used Calibrated Goals files
for each of the 35 countries to project average infections averted per year in the total population over a
10five-year period following a one-year temporary increase in PrEP coverage. Infections averted were
projected under two scenarios of ART coverage over five years: constant ART coverage and a 90-90-
90/95-95-95 scenario (percentage of the population who know their HIV status, percentage of those
testing HIV positive who are on ART, and percentage of those on ART who have achieved viral
suppression). The resulting estimates of infections averted per person-year on PrEP can be adjusted by
users.
Estimating the costs of PrEP required a new approach that incorporates a visit schedule to account for
high discontinuation rates and the different costs of initiation visits and continuation visits. While the
costs of the drugs used for PrEP were frequently known, default data were needed for the other
components of visit costs. We analyzed facility-level cost data to extract staff minutes per visit for
different priority populations. These data were combined with country-specific salary schedules to
estimate personnel costs, which are the foundation for calculating default values for other cost
components, including laboratory, overhead, capital, and above-site-level costs. Users can either use the
default values or enter their own cost data; these data are then linked to continuation rates and visit
schedules to estimate the annual cost per person initiating on and continuing to take PrEP.
For a more detailed cost analysis, users can elect to use the Detailed Costs module, which allows for
multiple “service delivery modes” (ways in which costs may vary, such as implementation model, type of
site, urban/rural, population served, subnational geographic area, implementer) with different cost
structures aimed at each population. A simpler Costs Lite module is also available for users with less
detailed costing needs or data available. Costs are specified in terms of personnel time, laboratory tests,
capital costs, and other recurrent costs (e.g., commodities, training) for initiation and continuation visits
separately, by population in Costs Lite and by both population and service delivery mode in Detailed
Costs. Above-site costs, such as demand creation or training, can be specified as annual lump sums by
population and service delivery mode. The schedule of continuation visits following initiation (e.g., every
month or every three months) can be specified for each population/service delivery mode; visit costs
are accrued over time according to visit schedules and subject to continuation rates. Cost results include
total cost per year, cost per HIV infection averted, and cost per person initiated, each available by
population and service delivery model.
PrEP-it contains an AGYW Geographic Prioritization tool that allows users to identify localized pockets
of high HIV risk among AGYW that are not captured by national- or provincial-level surveys. The user can
specify a list of sites, along with HIV prevalence (or test positivity rates) among AGYW, costs of PrEP and
ART, and other optional inputs for each area. A formula derived from a regression model estimates HIV
incidence from local prevalence among AGYW ages 15–24 (and, optionally, among AGYW ages 10–14,
for a better estimate).28 PrEP-it then identifies which catchment areas are cost-saving (where it is less
expensive to deliver PrEP than to pay for lifetime ART), or which catchment areas are cost-effective,
subject to a user-defined threshold for cost per infection averted.
The Drug Forecasting module builds on the infrastructure of the target-setting process, where users
select populations, enter continuation rates, and specify scale-up patterns. This module requires only
the number of pills distributed at initiation and continuation visits and the cost per pill. The tool then
calculates the number of ARVs needed each month and the estimated ARV costs.
11COUNTRY EXPERIENCES USING PrEP-IT
Since its release, PrEP-it has been downloaded more than 200 times. The governments of Eswatini,
Kenya, Tanzania, Zambia, and Zimbabwe have used PrEP-it as part of the target-setting process; the
experiences of Eswatini and Zimbabwe are documented in this paper. In addition, Vietnam and Ghana
used PrEP-it for drug forecasting, and their experiences are described below.
Eswatini
In Eswatini, PrEP is delivered to numerous priority populations, including SDCs, sex workers, transgender
persons, AGYW, MSM, and other men with increased likelihood of HIV acquisition. The Eswatini National
AIDS Programme, a part of the Ministry of Health, used the June 2019 PrEP-it field visit workshop as an
opportunity to set five-year national targets. During the workshop, participants discussed the various
questions the PrEP-it framework poses. For the priority populations, targets were set for SDCs, FSWs,
MSM, AGYW, transgender women, pregnant women, breastfeeding women, and a custom population of
men 30–34 years old. Continuation curves relied on demonstration project data for most priority
populations, but the team adjusted them to higher continuation rates for SDCs and pregnant women
based on the special needs of these populations. Default data for HIV prevalence and percentage
indicated for PrEP were used as a starting point for discussion, and three different S-shaped curves for
scale-up were assumed, allowing for variation in the peak scale-up month by priority population.
The group elected to set targets based on coverage, using the tool to estimate the number of initiations
needed to reach a specified coverage level at the end of the target period. The specified coverage levels
ranged from 10% for pregnant and breastfeeding women, transgender women, and AGYW ages 15–19
to 35% for SDCs. Based on these inputs, PrEP-it projected that a total of approximately 85,000 PrEP
initiations would be required over five years. Earlier in the target-setting process, higher coverage levels
had been selected, but they were adjusted downward after the group considered the feasibility of
implementation.
Zimbabwe
Zimbabwe’s national PrEP targets had been set for 2018–2020. In January–February 2020, the PrEP
Technical Working Group began setting new national PrEP targets for 2021–2023. The priority
populations included SDCs, FSWs, MSM, people who inject drugs (PWID), transgender persons, AGYW
ages 15–19 and 20–24, and pregnant and lactating women. The PrEP-it team used recent population size
estimates for SDCs, FSWs, and MSM. Where data were more uncertain, different population size
estimates for PWID and transgender persons were examined. Targets were set by coverage, ranging
from 4% for AGYW ages 15–19 to 24% for PWID, and different continuation rates were entered for each
priority population, based on partner data. An S-shaped scale-up pattern was selected for all
populations except SDCs, for whom a constant trend was assumed. Based on the different population
size estimates, the team presented two sets of targets to the PrEP Technical Working Group: 1) an
estimate of 126,000 initiations and 2) a more ambitious estimate of 142,000 initiations based on larger
population size estimates and a higher proportion of those estimated to be at elevated risk for HIV.
The PrEP Technical Working Group selected the 126,000 initiations targets as a more realistic goal to
achieve in the next three years.
12Ghana
In Ghana, PrEP delivery was aimed at SDCs, MSM, FSW, and transgender women; however, questions
arose as to whether PrEP provision might interfere with the ARV supply for HIV treatment. Ghana’s
national AIDS program was using tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) and
TDF/lamivudine (3TC) for HIV treatment and planned to use them for PrEP. To estimate the impact of
starting PrEP provision, the team used PrEP-it to estimate the quantity of ARVs that would be needed
each month to reach PEPFAR PrEP targets. The previously established targets for October 2020 through
September 2021 were entered into the Targets module, with 1,418 projected users over the period. In
addition, alternate scenarios of PrEP scale-up were estimated to determine the number of initiations
that would be needed each month to reach targets by the end of the fiscal year under different
scenarios.
The team estimated continuation rates using study and programmatic data from the peer-reviewed
literature and entered them in the Continuation module, with 55% of expected clients continuing at the
first month post-initiation and a gradual decline in continuation at month 3 (45%), month 6 (35%), and
one year (25%). An S-shaped curve with peak scale-up at month 6 was selected in the Targets module.
Based on these inputs, the team used PrEP-it to estimate the target number of people on PrEP each
month and, from that, the number of ARVs needed each month to fulfill the entered targets. The results
showed that PrEP provision would have a minimal impact on the ARV stock available at the national
level. As a result, it was determined that PrEP provision could commence.
Vietnam
Like Ghana, Vietnam used PrEP-it to estimate the number of ARVs needed, in this case as part of the
PEPFAR Country Operational Plan for 2020. The previously established target of 30,000, focusing on the
SDC, MSM, sex worker, PWID, and transgender populations, was entered into the Targets module. For
the Continuation module, the continuation rates were obtained from in-country programmatic data
showing relatively high levels of PrEP continuation after initiation: 95% at month 1, 89% at month 3, 76%
at month 6, and 69% at one year. An S-shaped curve was selected with a peak scale-up at three months.
Based on these inputs, PrEP-it was able to forecast the number of ARVs needed each month over 12
months from October 2020 through September 2021. As a result, PEPFAR Vietnam adjusted the amount
of ARVs originally planned for and shifted some of the funding that had been budgeted for ARVs to
efforts to drive demand for PrEP and increase the quality of PrEP service provision.
LESSONS LEARNED
Each step in the PrEP-it development process affirmed the need for an innovative tool to help users
navigate the myriad of challenges associated with PrEP program planning and implementation. PrEP-it
addresses these challenges by allowing users to set targets, track PrEP delivery, analyze patterns of PrEP
use, and estimate capacity, costs, drug supply needs, and impact in one accessible tool, ultimately
ensuring that planning meets the demand for PrEP commodities and is informed by concrete program
implementation at the service-delivery level.
The most important lesson learned from the tool development is the need to test the draft tool with
country partners to ensure that it meets their needs and is developed with the available data in mind.
Key insights were gleaned from each initial country visit and subsequent application of PrEP-it, resulting
13in tool adaptations and new modules, such as the Costs Lite, Drug Forecasting, and AGYW Geographic
Prioritization modules.
A key lesson from applying PrEP-it, observed in all the applications discussed here, is that continuation
rates have important implications for costs, target-setting, and drug forecasting. For costs, those
projecting budgetary needs for PrEP based on targets outside of PrEP-it often assumed PrEP
continuation rates far higher than the reality, and therefore projected high annual costs associated with
each client. The inclusion of more accurate assumptions about continuation rates produces a much
more realistic picture of the implementation costs and ARV needs associated with targets. For target-
setting, the inclusion of continuation rates and scale-up patterns allows users to estimate the number of
individuals on PrEP at any given point in time and understand how many PrEP initiations are needed to
reach desired coverage levels. In the applications, higher numbers of initiations were often needed than
were originally planned for due to discontinuation.
The ability to generate more accurate estimates of PrEP use also has important implications for drug
forecasting, enabling governments to predict their drug needs and not be constrained by unwarranted
fears that the introduction of PrEP will significantly affect the supply of ARV drugs for people living with
HIV who need life-long ART. In this sense, PrEP-it can be used as a tool to advocate for initiating PrEP
rollout.
Understanding coverage is particularly important in PrEP-it, where coverage means the proportion of
the eligible population taking PrEP at the end of the target period, not the proportion who have been
exposed to PrEP throughout the target period. By using coverage in the last month of the target period,
when programs have typically attained scale-up, PrEP-it users can focus on what is needed to achieve
epidemiological impact. During the application process, it became clear that low continuation rates
combined with high coverage goals led to unrealistically high targets. To achieve these targets, PrEP
clients would need to cycle on and off PrEP repeatedly during the year, and thus initiation targets far
exceeded the number of potential users. Therefore, we recommend setting target coverage levels below
the specified six-month continuation rates. If higher coverage levels are desired, planners and
implementers may wish to use aspirational continuation rates, with the understanding that increasing
continuation is critical to achieving program goals. Alternatively, if PrEP users are cycling on and off PrEP
multiple times per year and program planners want to capture that dynamic, they need to understand
that the number of initiations (which include reinitiations) can reasonably exceed the population size.
An additional lesson is the need to set aside sufficient time for applying PrEP-it. Time is required for
internal discussions about which PrEP-it functions, data sources, and assumptions should be used. In this
sense, PrEP-it serves as an organizing platform that
can help program planners and implementers who PrEP-it serves an organizing platform that
have different priorities and foci work together to can help program planners and
better plan for and provide PrEP; however, multiple implementers work together to better
meetings are typically required for these plan for and provide PrEP.
negotiations before targets can be finalized.
As with every tool, PrEP-it has limitations. First, it does not address every planning and management
need for PrEP programs. For example, several in-country stakeholders expressed a need for assistance
with demand creation tools. PrEP-it can indicate when demand creation is needed by identifying where
the ability to reach targets may be limited by demand, but it does not provide guidance on demand
generation. Second, although PrEP-it provides default data inputs for some variables, they are often
14based on assumptions or data aggregated at the national level; users should tailor inputs to their specific
program contexts. Third, PrEP-it does not address PrEP cycling, when individuals start, stop, and restart
PrEP, or event-driven PrEP, when PrEP is used for a specific exposure. Fourth, the current version of PrEP
is limited to oral PrEP, but a forthcoming version will be expanded to incorporate a method mix of oral
and vaginal ring PrEP.
Finally, although Microsoft Excel is readily available to most potential users, version control and
difficulty managing updates are concerns. In addition, the two-dimensional nature of Excel makes it
difficult to easily capture the multidimensionality of PrEP programs. A forthcoming software version of
PrEP-it, planned for release in 2021, will allow for a more user-friendly interface and additional features,
including planning for and tracking multiple PrEP modalities, such as oral, vaginal ring, and injectable
PrEP.
CONCLUSION
Prior to PrEP-it, few PrEP planning and M&E tools were available beyond the existing systems that track
PrEP initiations for PEPFAR reporting purposes. PrEP-it’s increased focus on continuation rates, scale-up,
and population coverage provides an important advancement for PrEP planning and management. In
the coming years, PrEP programming will undoubtedly evolve, and PrEP-it is designed to co-evolve to
enhance successful scale-up of PrEP. In the meantime, the current version of PrEP-it and the
accompanying user guide are available at no cost to any program interested in using it to help improve
planning, monitoring, and evaluation of the delivery of PrEP.
15REFERENCES
1. World Health Organization (WHO). WHO expands recommendation on oral preexposure prophylaxis
of HIV infection (PrEP). Policy Brief. WHO; 2015. https://www.who.int/hiv/pub/prep/policy-brief-
prep-2015/en/. Accessed 21 May 2021.
2. AVAC. PrEPWatch: Global PrEP Tracker. https://www.prepwatch.org/resource/global-prep-tracker/.
Accessed 21 May 2021.
3. Hodges-Mameletzis I, Dalal S, Msimanga-Radebe B, Rodolph A, Baggaley R. Going global: the
adoption of the World Health Organization’s enabling recommendation on oral pre-exposure
prophylaxis for HIV. Sex Health. 2018;15(6): 489-500.
4. Uganda Ministry of Health. Consolidated guidelines for prevention and treatment of HIV in Uganda.
Uganda Ministry of Health; 2020.
https://uac.go.ug/sites/default/files/Consolidated%20HIV%20Guidelines%202020.pdf. Accessed 21
May 2021.
5. Lesotho Ministry of Health. National guidelines on the use of antiretroviral therapy for HIV
prevention and treatment. 5th ed. Lesotho Ministry of Health; 2016.
https://www.childrenandaids.org/sites/default/files/2017-04/Lesotho_ART-Guidelines_2016.pdf.
Accessed 1 Feb 2021.
6. Prevention gap report. UNAIDS; 2016.
https://www.unaids.org/en/resources/documents/2016/prevention-gap. Accessed 21 May 2021.
7. Coleman R. Setting the scene, setting the targets. The Joint United Nations Programme on HIV/AIDS
prevention targets of 2016 and estimating global pre-exposure prophylaxis targets. Sex Health.
2018;15(6): 485-88.
8. O’Malley G, Barnabee G, Mugwanya K. Scaling-up PrEP delivery in sub-Saharan Africa: What can we
learn from the scale-up of ART? Cur HIV/AIDS Rep. 2019;16: 141-50.
9. Pyra MN, Haberer JE, Hasen N, Reed J, Mugo NR, Baeten JM. Global implementation of PrEP for HIV
prevention: setting expectations for impact. J Int AIDS Soc. 2019;22: e25370.
10. Pretorius C, Schnure M, Dent J, Glaubius R, Mahiane G, Mitchell G, et al. Modelling impact and cost-
effectiveness of oral pre-exposure prophylaxis in 13 low-resource countries. J Int AIDS Soc.
2020;23(2): e25451.
11. AVAC. Ongoing and planned PrEP demonstration and implementation studies. AVAC.
https://www.avac.org/resource/ongoing-and-planned-prep-demonstration-and-implementation-
studies. Accessed 20 Oct 2019.
12. Irungu EM, Sharma M, Maronga C, Mugo N, Ngure K, Celum C, et al. The incremental cost of
delivering PrEP as a bridge to ART for HIV serodiscordant couples in public HIV care clinics in Kenya.
AIDS Res Treat. 2019 May 2;4170615. doi: 10.1155/2019/4170615.
13. Roberts DA, Barnabus RV, Abuna F, Lagat H, Kinuthia J, Pintye J, et al. The role of costing in the
introduction and scale-up of HIV pre-exposure prophylaxis: evidence from integrating PrEP into
routine maternal and child health and family planning clinics in western Kenya. J Int AIDS Soc.
2019;22(Suppl 4): e25296. doi: 10.1002/jia2.25296.
14. Wamoni E. Cost of PrEP delivery in public sector HIV care clinics in Kenya. Partners Scale-up Project;
2019.
15. Kioko U, Forsythe S. Estimating the costs of pre-exposure prophylaxis (PrEP) in ten counties in
Kenya. Jilinde Project; 2019.
16. Jamieson L, Meyer-Rath G. PrEP Cost Model South Africa, version 5.2. Health Economics and
Epidemiology Research Office (HE2RO), Wits Health Consortium, University of the
Witwatersrand/Department of Global Health and Boston University; February 2019.
1617. Hendrickson C, Long L, van Rensburg C, van de Vijver D, Claassen C, Njelesani M, et al. Zambia PrEP
program and service delivery scale-up: Comparison of service delivery projects. EQUIP; 2019.
18. Magnenah C, Nhamo D, Terris-Prestholt F. The costs of PrEP implementation in seven Zimbabwe
clinics. Durham, NC, USA: OPTIONS Consortium; 2019.
19. Bollinger LA, Sanders S, Winfrey W, Adesina A. Lives Saved Tool (LiST) costing: a module to examine
costs and prioritize interventions. BMC Public Health. 2017;17(Suppl 4): 782.
20. Evans DB, Edejer TT, Adam T, Lim SS. Methods to assess the costs and health effects of interventions
for improving health in developing countries. BMJ. 2005;331(7525): 1137-40.
21. Avenir Health Spectrum Model, version 5.761. Avenir Health. http://avenirhealth.org/software-
spectrum.php. Accessed 16 Oct 2019.
22. Eakle R, Gomez GB, Naiker N, Bothma R, Mbogua J, Cabrera Escobar M, et al. HIV pre-exposure
prophylaxis and early antiretroviral treatment among female sex workers in South Africa: results
from a prospective observational demonstration project. PLoS Med. 2017;14(11): e1002444.
23. Celum C, Travill D, Baeten J, Momollo O. POWER Prevention Options for Women Evaluation
Research: PrEP implementation for AGYW – lessons being learned. PowerPoint presentation.
November 14, 2018.
24. Kyongo JK. How long will they take it? Oral pre-exposure prophylaxis (PrEP) retention for female sex
workers, men who have sex with men and young women in a demonstration project in Kenya. 22nd
International AIDS Conference; 2018 Jul 23-27; Amsterdam, The Netherlands.
25. Stevens O. Technical updates and Naomi model development report and recommendations from a
meeting of the UNAIDS Reference Group on Estimates, Modelling, and Projections, Montreux,
Switzerland - 8-10th October 2019. Geneva: UNAIDS, 2019.
http://www.epidem.org/sites/default/files/reports/Tech%20Updates%20Report_final.pdf. Accessed
21 May 2021.
26. Stankevitz K, Grant H, Lloyd J, Gomez GB, Kripke K, Torjesen K, et al. Oral preexposure prophylaxis
continuation, measurement and reporting. AIDS. 2020;34(12): 1801-11.
27. Stover J, Bollinger L, Izazola JA, Loures L, DeLay P, Ghys PD, et al. What is required to end the AIDS
epidemic as a public health threat by 2030? The cost and impact of the fast-track approach. PloS
One. 2016;11(5): e0154893.
28. Hamilton M, Mahiane G, Pretorius C, Kripke K. Simple tool for geographic prioritization of PrEP for
adolescent girls and young women based on cost-effectiveness. HIV R4P Virtual Conference; 2021
Jan 27–28 and 3–4 Feb. https://programme.hivr4p.org/Abstract/Abstract/760. Accessed 30 April
2021.
17You can also read
